Fuzzy Pattern Recognition Based on Generalized Euclidean Weight Distance Adjoined Degree and Its Application in Forecasting Hazard of Karst Collapse

Abstract

The karst collapse is very complicated and its influential factors are various and interdependent, besides, the geologic body itself is heterogeneous and uncertain.The fuzzy pattern recognition,which is a new fuzzy recognition method and can make the sorting more exactly, has been used in many territories.Therefore, it is necessary to adopt the fuzzy pattern recognition to forecast harzard of karst collapse.Because the existing adjoined degree couldn’t recognize efficiently, weight value was mingled with adjoined degree, then the fuzzy pattern recognition based on generalized euclidean weight distance adjoined degree was made in this paper, the reasonableness for the application of this model to forecast karst collpase hazard was discussed. Taking the karst collpase in Wuhan city as an example, 13 influence factors were chose, based on the new fuzzy pattern recognition method, the harzard of karst collapse in Wuhan city was forecast, it was found that the forecasting result was pretty accurate. Therefore, it could be concluded that the applicaition of fuzzy pattern recognition based on generalized eclidean weight distance adjoined degree in forecasting hazard of karst collpase was effective and reasonable.